Bootstrap Functions for Bivariate POT
Obtains bootstrap samples for the bivariate POT model as output from fbvpot.
bvpotbooter(object, B, rsize, block.length = 1, shuffle = NULL, replace = TRUE, ...)
object |
A list object of class “fbvpot”. |
B |
integer giving the number of bootstrap resamples to obtain. |
rsize |
integer giving the size of the bootstrap samples. Defaults to the (original) sample size. For heavy-tailed data, may want to use a smaller size (e.g., square root of the original sample size). |
block.length |
integer giving the length of blocks to use (for dependent data). Default is unity, which yields IID samples. |
shuffle |
(optional) |
replace |
logical, should the resamples be taken with replacement? |
... |
Not used. |
Performs bootstrap resampling for bivariate POT inference. CI's are obtained by calling the ci function. See the help file for booter from the distillery package for more details.
A list object of class “booted”.
call |
the function call |
data |
original data series |
statistic, B, block.length, replace, rsize |
same arguments as passed in |
results |
B column matrix giving the paramter estimates for each bootstrap resample. |
type |
character stating whether the resample method is iid or cbb. |
Eric Gilleland
data( "SantaAna" ) Z <- SantaAna[,3:4] mfit1 <- fevd( x = temp, data = Z, threshold = 36.75, type = "GP" ) mfit2 <- fevd( x = windspeeds, data = Z, threshold = 7.09875, type = "GP" ) fit2 <- fbvpot( x = Z, threshold = apply( Z, 2, quantile, probs = 0.95 ), tform = "tf", fit = list( mfit1, mfit2 ) ) # Note, using B = 50 below for R check speed, but probably need # a higher number. fit2booted <- bvpotbooter( fit2, B = 50, rsize = 62, block.length = 7 ) summary( fit2booted ) ci( fit2booted, type = "perc" )
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.